David Zhang

An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models. The goal of this repo is to provide the simplest implementation of a deep research agent - e.g. an agent that can refine its research direction overtime and deep dive into a topic.

Reusable AI agent skills for software factories: explore ideas, write specs, implement, review, and run autonomous research. Works with Claude Code, Codex, and other skill-compatible agents.

Run Claude Agent (Claude Code) in a sandbox, control it via websocket

An open-source, headless CRM built for agents. No UI. No dashboard. Just a CLI and a FUSE-mounted filesystem.

Help the Fly Escape — MaleCNS connectome LIF tech demo (whimsical house escape)

An opinionated full-stack agent harness with native memories, long running tasks, and multi-agent relay

A TypeScript library providing OAuth2 authentication utilities for Model Context Protocol (MCP) clients. This library simplifies the process of adding OAuth authentication to MCP client implementations.

Simple API for creating a webhook endpoint to capture webhook requests. Built for AI agents that cannot receive pushed webhooks, this allows agents to pull webhook events as needed.

Local, Git-native evidence and review for Codex and Claude Code sessions

CLI-first photo library + editor for Mac (portable core)

How LLMs work, from first principles — an interactive 3D explainer

Ian-style Xiaohei editorial illustrations, English version — hand-drawn article images with English annotations